A reframing within Resilience Frontiers’ Pathway 3: Universal and Equitable Access to Frontier Technologies
the very tools designed to safeguard our planet are themselves resource-intensive, with data centers projected to consume 3-4% of global power demand by 2030, and training a single large language model producing carbon emissions equivalent to dozens of transatlantic flights.
This duality underscores a fundamental question: Whose vision governs AI’s role in climate strategies? Without intentional governance, frontier technologies risk widening existing inequalities—particularly for Indigenous communities, smallholder farmers, and coastal populations already bearing the brunt of climate impacts. The path forward requires what we term “endogenization”: embedding technological evolution within socio-ecological systems rather than treating it as an external force imposed upon vulnerable communities.
From Prediction to Proactive Stewardship
Current AI applications in climate action remain predominantly reactive—predicting floods, forecasting droughts, and responding to disasters after they unfold. While valuable, this approach constrains us to known scenarios and historical data, leaving communities ill-prepared for unprecedented climate events.
A proactive paradigm shifts the question from “Where will AI take us?” to “How do we actively co-steer its development?” This requires moving beyond optimization to what we call prospective stewardship—using frontier technologies not merely to predict climate futures but to envision and create new pathways for human-Earth coexistence. Technologies like big data, AI, digital twinning platforms, and cutting-edge Earth observation systems become tools for collective imagination, not just calculation.
The Endogenization Framework
To ensure frontier technologies serve equitable climate resilience, we propose a circular framework for their integration:
Reduction of Hazard: AI-enhanced risk assessment and early warning systems—such as the Philippines’ Project NOAH for flood prediction—protect lives and livelihoods. Yet these systems must be redesigned with structural vulnerabilities in mind, ensuring that resource inequality and information-access gaps are addressed, not reinforced.
Continual Learning and Adaptation: AI transcends harm reduction to enable regenerative futures. In agriculture, AI-driven platforms like India’s Krishi Vigyan Kendra provide climate-resilient crop advisory services, while the Coordinated Regional Climate Downscaling Experiment (CORDEX) integrates Indigenous ecological knowledge with meteorological data—ensuring community agency over passive risk prediction.
Ongoing Reassessment: Continuous evaluation ensures socio-economic equity as deployments scale. Post-implementation reflection requires technical expertise alongside evolving community needs, government investments, and accessible infrastructure—ensuring technologies serve those most vulnerable to climate injustice
Governance for Collective Benefit
Frontier technologies governed ethically in the collective interests of people and planet can safeguard the global commons while providing tools for sustainable abundance and circularity. This demands three governance shifts:
Equity Impact Assessments: Required for all AI-climate projects, evaluating effects on marginalized communities’ access to water, land, and decision-making power.
Community Data Trusts: Enabling Indigenous groups and local communities to govern environmental data collection and use, drawing inspiration from Māori data sovereignty principles.
Open-Source Climate AI Platforms: Mandating public funding for transparent, accessible tools that grassroots organizations can adapt to local contexts.
The Last Mile: Where Implementation Meets Vision
The true test of frontier technologies lies in the “last mile”—where technical solutions meet real communities facing real climate impacts. AI-enhanced irrigation systems mean little if smallholder farmers cannot access them. Sophisticated climate models serve no purpose if Indigenous communities’ place-based knowledge is excluded.
The UK’s Energy Data Task Force and Germany’s “AI for Environment and Climate” program demonstrate initial steps toward endogenization—making data and technology accessible within policy frameworks. Yet full endogenization requires a cultural shift toward cross-sectoral collaboration, where experimentation and collective problem-solving flourish with AI as a central enabler, not a replacement for human wisdom.
The goal is not a “final solution” but continuous co-adaptation as planetary conditions change. By anchoring AI-climate strategies in enforceable, rights-based frameworks—rather than voluntary pledges—we can ensure technologies serve systemic equity, not just efficiency. The question of who decides over what is decided determines whether frontier technologies become bridges to resilience or accelerants of inequality.
Frontier technologies offer us the tools. It is our collective responsibility to ensure they serve humanity’s most vulnerable—and the living systems upon which all life depends.
Reference:
This article is adapted from the original publication:
Zheng, Y., Jin, Y., Nassef, Y., Fang, D., Strohmayer, U., & Zhao, Z. (2025). “Embedding the frontiers in climate action: navigating the evolving earth-human ecosystem.” npj Climate Action, 4:98.
DOI: https://doi.org/10.1038/s44168-025-00304-8
For full academic citations and detailed analysis, please refer to the original publication